Skip to content
Agentmart
Submit a tool

Write Impact Summary

Turn work evidence into concise Situation-Task-Action-Result-Impact narratives without inventing outcomes.

Skill by YuiZhou

IndexedMIT★ 2
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/yuizhou-write-impact-summary

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Turn work evidence into concise Situation-Task-Action-Result-Impact narratives without inventing outcomes.

Use cases: writing, productivity

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/write-impact-summary

What's inside

Skill name
Write Impact Summary
Description
Turn work evidence into concise Situation-Task-Action-Result-Impact narratives without inventing outcomes.
License
MIT
Instructions
66 lines in SKILL.md · view on GitHub

Agent Skills spec issues

  • name: only lowercase letters, digits and single hyphens allowed
  • name: "Write Impact Summary" does not match folder "dayarc-write-impact-summary"

Files (1)

  • SKILL.md

SKILL.md (first lines)

## Input

```json
{
  "period": "daily | weekly | monthly",
  "signals": [],
  "themes": [],
  "activities": [],
  "accomplishments": [],
  "prior_impact_summaries": []
}
```

Inputs may come directly from collected signals for a daily summary, from daily profiles for a weekly summary, or
from weekly summaries for a monthly summary.

## Output (JSON)

```json
{
  "impact_summaries": [
    {
      "title": "Hardened GitHub brief delivery",
      "situation": "Brief delivery could resolve the wrong repository when another GitHub account was active.",
      "task": "Make repository selection authoritative across multi-account environments.",
      "action": "Added account-aware repository resolution and explicit failure when the configured repository is unavailable.",
      "result": "Briefs now stop instead of being redirected to a different repository.",
      "impact": "Protects private work summaries from being delivered to the wrong audience.",
      "status": "completed",
      "source_breadcrumbs": ["github.com/owner/repo/pull/123"]
    }
  ]
}
```

`result` and `impact` are optional when the available evidence does not support them.

## Instructions

1. Build summaries around meaningful outcomes, decisions, risks reduced, people unblocked, or progress toward a
   goal. Do not create one summary per email, commit, repository, meeting, or chat thread.
2. Merge related evidence into one narrative even when it comes from different connectors or repositories.
3. Use a compact STAR-style structure:
   - `situation` — the problem, need, constraint, or goal that made the work relevant.
   - `task` — the responsibility or intended change.
   - `action` — the concrete work the user performed. Use strong verbs and name the object changed.
   - `result` — an outcome directly supported by evidence, such as merged, shipped, approved, resolved, validated,
     documented, decided, or unblocked.
   - `impact` — why the result matters to users, reliability, delivery, cost, security, quality, or team velocity.
4. Never invent metrics, business impact, completion, causality, or stakeholder benefit. If evidence only shows
   activity, omit `result` and `impact`, set `status` to `"in_progress"`, and make the action describe concrete
   progress rather than claiming success.
5. Use `status` values:
   - `completed` — a completion or terminal result is explicit in the evidence.
   - `advanced` — a review, decision, milestone, or meaningful intermediate result is explicit.
   - `in_progress` — work occurred but no result is yet observable.
6. Every summary must include at least one `source_breadcrumbs` entry. Deduplicate breadcrumbs and retain the most
   useful links, issue/PR IDs, meeting titles, or thread subjects. Maximum 4 breadcrumbs per summary.
7. Prefer 2–5 summaries for daily and weekly periods and 3–6 for monthly periods. Rank by observed impact,
   completion, effort, and strategic relevance.
8. For weekly and monthly rollups, synthesize across prior `impact_summaries`; do not merely concatenate or repeat
   daily/weekly wording. Describe the larger arc and deduplicate the same outcome across periods.
9. Keep each field to one concise sentence. The full summary should be understandable without opening the source,
   while the breadcrumbs preserve traceability.

From YuiZhou/dayarc-agent, MIT.

More from YuiZhou/dayarc-agent

Signals

Stars
2
Forks
2
Open issues
7
Last commit (this folder)
Aug 26, 2026 (2 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from YuiZhou

Cite this page

Write Impact Summary skill. Agent Mart. https://agentmart.dev/skills/yuizhou-write-impact-summary (last updated 2026-10-24)